U.S. patent application number 17/031897 was filed with the patent office on 2022-03-03 for non-contact motion detection method, motion detection device and emergency detection method.
This patent application is currently assigned to Wistron Corporation. The applicant listed for this patent is Wistron Corporation. Invention is credited to Kuo-Hsing Wang, Chih-Ping Wu.
Application Number | 20220066016 17/031897 |
Document ID | / |
Family ID | 1000005138644 |
Filed Date | 2022-03-03 |
United States Patent
Application |
20220066016 |
Kind Code |
A1 |
Wang; Kuo-Hsing ; et
al. |
March 3, 2022 |
NON-CONTACT MOTION DETECTION METHOD, MOTION DETECTION DEVICE AND
EMERGENCY DETECTION METHOD
Abstract
A non-contact motion detection method, a motion detection device
and an emergency detection method are provided. The non-contact
motion detection method includes: receiving a reflection signal
from a field to obtain a raw data signal; determining that a first
event occurs in the field according to an energy value of the raw
data signal, and providing a first alarm; determining that a second
event occurs in the field according to an energy distribution of
the reflection signal; and in case of determining that the second
event occurs, providing a second alarm corresponding to the second
event according to the energy value of the raw data signal.
Inventors: |
Wang; Kuo-Hsing; (New Taipei
City, TW) ; Wu; Chih-Ping; (New Taipei City,
TW) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
Wistron Corporation |
New Taipei City |
|
TW |
|
|
Assignee: |
Wistron Corporation
New Taipei City
TW
|
Family ID: |
1000005138644 |
Appl. No.: |
17/031897 |
Filed: |
September 25, 2020 |
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
G01S 13/583 20130101;
G01S 7/352 20130101; G01S 13/62 20130101 |
International
Class: |
G01S 13/62 20060101
G01S013/62; G01S 13/58 20060101 G01S013/58; G01S 7/35 20060101
G01S007/35 |
Foreign Application Data
Date |
Code |
Application Number |
Sep 1, 2020 |
TW |
109129816 |
Claims
1. A non-contact motion detection method, comprising: transmitting
a detection signal to a field, and receiving a reflection signal
corresponding to the field; processing the reflection signal to
obtain a raw data signal; calculating a first maintenance time
length corresponding to an energy value of the raw data signal
continuously being less than a first preset energy value; in
response to the first maintenance time length being greater than or
equal to a first preset time length, determining that a first event
occurs in the field, and providing a first alarm corresponding to
the first event; determining whether a second event occurs in the
field according to an energy distribution of the raw data signal;
in response to the second event that occurs, calculating a second
maintenance time length corresponding to the energy value of the
raw data signal continuously being less than a second preset energy
value; and in response to the second maintenance time length being
greater than or equal to a second preset time length, providing a
second alarm corresponding to the second event.
2. The non-contact motion detection method of claim 1, wherein
before calculating the energy value of the raw data signal, the
method further comprises filtering the raw data signal based on a
frequency range.
3. The non-contact motion detection method of claim 1, wherein the
step of calculating the first maintenance time length corresponding
to the energy value of the raw data signal continuously being less
than the first preset energy value comprises: when the energy value
of the raw data signal is less than the first preset energy value,
timing a maintenance time during which an energy value of the
reflection signal is less than the first preset energy value to
generate the first maintenance time length.
4. The non-contact motion detection method of claim 3, wherein the
step of calculating the first maintenance time length corresponding
to the energy value of the raw data signal continuously being less
than the first preset energy value further comprises: when the
energy value of the reflection signal is greater the first preset
energy value, resetting the first maintenance time length.
5. The non-contact motion detection method of claim 1, wherein in
response to the second event that occurs, the step of calculating
the second maintenance time length corresponding to the energy
value of the raw data signal continuously being less than the
second preset energy value comprises: in case of determining that
the second event occurs, timing a time length during which the
energy value of the raw data signal is continuously less than the
second preset energy value to generate the second maintenance time
length.
6. The non-contact motion detection method of claim 5, further
comprising: in case of determining that the second event occurs,
timing a time length during which the energy value of the raw data
signal is continuously greater than the second preset energy value
to generate a third maintenance time length; and when the third
maintenance time length is less than a third preset time length,
resetting the third maintenance time length.
7. The non-contact motion detection method of claim 6, further
comprising: in case of determining that the second event occurs,
stopping timing when the third maintenance time length is greater
than or equal to the third preset time length.
8. The non-contact motion detection method of claim 1, wherein the
first event is an event in which a fast motion does not occur in
the field within the first preset time length.
9. The non-contact motion detection method of claim 1, wherein the
second event is a fall event that occurs in the field.
10. The non-contact motion detection method of claim 1, wherein the
step of determining whether the second event occurs in the field
according to the energy distribution of the raw data signal further
comprises: analyzing the energy distribution of the raw data signal
according to a motion detection algorithm, wherein the motion
detection algorithm is trained by an artificial neural network.
11. A motion detection device, comprising: a detection circuit,
transmitting a detection signal to a field, receiving a reflection
signal corresponding to the field, and processing the reflection
signal to obtain a raw data signal; and a processor, coupled to the
detection circuit, and configured to: calculate a first maintenance
time length corresponding to an energy value of the raw data signal
continuously being less than a first preset value; when the first
maintenance time length is greater than or equal to a first preset
time length, determine that a first event occurs in the field, and
provide a first alarm corresponding to the first event; determine
that a second event occurs in the field according to an energy
distribution of the raw data signal; in case of determining that
the second event occurs, calculate a second maintenance time length
corresponding to the energy value of the raw data signal
continuously being less than a second preset value; and when the
second maintenance time length is greater than or equal to a second
preset time length, provide a second alarm corresponding to the
second event.
12. The motion detection device of claim 11, wherein before
calculating the energy value of the raw data signal, the processor
performs a filtering operation on the raw data signal based on a
frequency range.
13. The motion detection device of claim 11, wherein the processor
comprises: a first timer, configured to, when the energy value of
the raw data signal is less than the first preset energy value,
accumulate a maintenance time during which the energy value of the
raw signal is less than the first preset energy value to generate
the first maintenance time length.
14. The motion detection device of claim 13, wherein the processor
further comprises: a second timer, configured to, in case of
determining that the second event occurs, accumulate a time length
during which the energy value of the raw data signal is
continuously less than the second preset energy value to generate
the second maintenance time length.
15. The motion detection device of claim 14, wherein the processor
further comprises: a third timer, configured to, in case of
determining that the second event occurs, timing a time length
during which the energy value of the raw data signal is
continuously greater than the second preset energy value to
generate a third maintenance time length, wherein when the
processor determines that the third maintenance time length is less
than a third preset time length, the processor resets the third
timer.
16. The motion detection device of claim 15, wherein in case of
determining that the second event occurs, the processor instructs
the third timer to stop timing when determining that the third
maintenance time length is greater than or equal to the third
preset time length.
17. The motion detection device of claim 11, wherein the first
event is an event in which a fast motion does not occur in the
field within the first preset time length.
18. The motion detection device of claim 11, wherein the second
event is a fall event that occurs in the field.
19. The motion detection device of claim 11, wherein the motion
detection device is a continuous wave radar, and the detection
circuit is a radio frequency transceiver.
20. An emergency detection method, comprising: receiving a
reflection radio frequency signal, and processing the reflection
radio frequency signal to obtain a raw data signal; processing the
raw data signal to obtain an energy response corresponding to a
field, and determining a first detection result according to the
energy response; in response to the energy response matching a
first event, determining whether the first detection result belongs
to an emergency; processing the raw data signal according to a
motion detection algorithm to obtain a second detection result; and
in response to the second detection result matching a first motion,
determining whether the second detection result belongs to the
emergency.
Description
CROSS-REFERENCE TO RELATED APPLICATION
[0001] This application claims the priority benefit of Taiwan
application no. 109129816, filed on Sep. 1, 2020. The entirety of
the above-mentioned patent application is hereby incorporated by
reference herein and made a part of this specification.
TECHNICAL FIELD
[0002] The disclosure relates to a motion detection method, a
motion detection device, and an emergency detection method, and
more particularly to a non-contact detection technology that
improves a recognition rate for emergency.
BACKGROUND
[0003] In the existing motion detection technology, whether a
detection object (e.g., elderly, patient or child) has fallen,
slipped or collided can be determined by the motion detection
technology based on a fast motion of the detection object so that
an alarm can be provided accordingly to inform a caregiver of the
detection object. Therefore, how to improve the recognition rate of
recognition rate for emergency is one of the research topics of
those skilled in the art.
SUMMARY
[0004] The disclosure provides a non-contact motion detection
method and a motion detection device, which can detect a field
where a detection object is located and energy of motions of the
detection object, so as to improve the accuracy of the situation in
the field when the detection object is in danger, and provide the
correct alarm.
[0005] An embodiment of the disclosure discloses a non-contact
motion detection method. The non-contact motion detection method
includes: transmitting a detection signal to a field, and receiving
a reflection signal corresponding to the field; processing the
reflection signal to obtain a raw data signal; calculating a first
maintenance time length corresponding to an energy value of the raw
data signal continuously being less than a first preset energy
value; in response to the first maintenance time length being
greater than or equal to a first preset time length, determining
that a first event occurs in the field, and providing a first alarm
corresponding to the first event; determining whether a second
event occurs in the field according to an energy distribution of
the reflection signal; in response to the second event that occurs,
calculating a second maintenance time length corresponding to the
energy value of the raw data signal continuously being less than a
second preset energy value; and in response to the second
maintenance time length being greater than or equal to a second
preset time length, providing a second alarm corresponding to the
second event.
[0006] An embodiment of the disclosure discloses a motion detection
device. The motion detection device includes a detection circuit
and a processor. The detection circuit transmits a detection signal
to a field, receives a reflection signal corresponding to the
field, and processes the reflection signal to obtain a raw data
signal. The processor is connected to the detection circuit. The
processor calculates a first maintenance time length corresponding
to an energy value of the raw data signal continuously being less
than a first preset energy value, and when the first maintenance
time length is greater than or equal to a first preset time length,
determines that a first event occurs in the field, and provides a
first alarm corresponding to the first event. The processor further
determines that a second event occurs in the field according to an
energy distribution of the raw data signal, in case of determining
that the second event occurs, calculates a second maintenance time
length corresponding to the energy value of the raw data signal
continuously being less than a second preset value, and when the
second maintenance time length is greater than or equal to a second
preset time length, provides a second alarm corresponding to a
second event.
[0007] An embodiment of the disclosure discloses an emergency
detection method. The emergency detection method includes:
receiving a reflection radio frequency signal, and processing the
reflection radio frequency signal to obtain a raw data signal;
processing the raw data signal to obtain an energy response
corresponding to a field, and determining a first detection result
according to the energy response; in response to the energy
response matching a first event, determining whether the first
detection result belongs to an emergency; processing the raw data
signal according to a motion detection algorithm to obtain a second
detection result; and in response to the second detection result
matching a first motion, determining whether the second detection
result belongs to the emergency.
[0008] Based on the above, the disclosure receives the reflection
signal from the field and processes the reflection radio frequency
signal to obtain the raw data signal. When the first maintenance
time length is greater than or equal to the first preset time
length, it is determined that the first event occurs in the field,
and the first alarm is provided. The disclosure further determines
that the second event occurs in the field according to the energy
distribution of the raw data signal. In case of determining that
the second event occurs, the second alarm is provided when the
second maintenance time length is greater than or equal to the
second preset time length. In this way, the disclosure can detect
the energy generated by the motion of the detection object in the
sensing field, so as to improve the accuracy of the situation in
the field when the detection object is in danger, and provide the
correct alarm.
BRIEF DESCRIPTION OF THE DRAWINGS
[0009] FIG. 1 is a schematic diagram illustrating a motion
detection device and a field according to an embodiment of the
disclosure.
[0010] FIG. 2 is a schematic diagram illustrating a motion
detection device according to an embodiment of the disclosure.
[0011] FIG. 3 is a schematic diagram illustrating a motion
detection device according to an embodiment of the disclosure.
[0012] FIG. 4 is a flowchart illustrating a non-contact motion
detection method according to an embodiment of the disclosure.
[0013] FIG. 5A to FIG. 5C are flowcharts illustrating a non-contact
motion detection method according to an embodiment of the
disclosure.
[0014] FIG. 6 is a schematic diagram illustrating a processor
according to an embodiment of the disclosure.
[0015] FIG. 7 is a schematic diagram illustrating a conversion of
data of a reflection signal into an energy window slot according to
an embodiment of the disclosure.
DETAILED DESCRIPTION
[0016] Referring to FIG. 1, FIG. 1 is a schematic diagram
illustrating a motion detection device and a field according to an
embodiment of the disclosure. A motion detection device 10 provides
a detection signal SD to a field DR, and receives a reflection
signal SR from the field DR. The reflection signal SR is, for
example, a reflection radio frequency signal. The energy of the
reflection signal SR includes the energy generated by any motion in
the field DR. The field DR may be at least a part of any indoor
space. In this embodiment, the field DR can provide the reflection
signal SR corresponding to any motion in the field DR based on the
detection signal SD. The motion detection device 10 performs a
signal processing on the reflection signal to obtain a raw data
signal SRD. The motion detection device 10 then determines whether
a first event occurs on a detection object U in the field DR
according to the raw data signal SRD, and accordingly provides a
first alarm ALM1 corresponding to the first event. In addition, the
motion detection device 10 further determines whether a second
event occurs on the detection object U in the field DR according to
an energy distribution of the raw data signal SRD in time domain,
and determines whether to provide a second alarm ALM2 corresponding
to the second event according to the energy of the raw data signal
SRD when the second event occurs.
[0017] Referring to FIG. 1, FIG. 2 and FIG. 4 together, FIG. 2 is a
schematic diagram illustrating a motion detection device according
to an embodiment of the disclosure. FIG. 4 is a flowchart
illustrating a non-contact motion detection method according to a
first embodiment of the disclosure. In this embodiment, a motion
detection device 20 includes a detection circuit 210 and a
processor 220. The non-contact motion detection method can be
applied to the motion detection device 10. The motion detection
devices 10 and 20 may be a continuous wave radar. In another
embodiment, the motion detection devices 10 and 20 may be a
frequency modulated continuous wave radar. The detection circuit
210 includes a radio frequency transceiver and a radio frequency
signal processing circuit. In step S110, the detection circuit 210
transmits the detection signal SD to the field DR, and receives the
reflection signal SR from the field DR. The field DR may be at
least a part of any indoor space. In some embodiments, multiple
motion detection devices 10 may be provided to detect a movement
state and physiological information of the detection object U in an
indoor space. In step S120, the detection circuit 210 performs a
signal processing on the reflection signal, and then obtains the
reflection signal SRD based on a sampling frequency. In this
embodiment, the sampling frequency is 500 Hz (but the disclosure is
not limited thereto).
[0018] In step S130, the processor 220 is connected to the
detection circuit 210 to receive the raw data signal SRD and
calculate an energy value of the raw data signal SRD. The processor
220 also filters the raw data signal SRD based on a specific
frequency range. The filtering operation may be performed before
step S130. In this embodiment, the specific frequency range is
between 100 Hz and 250 Hz. The specific frequency range includes a
frequency range corresponding to a fast movement or a fast motion
(e.g., walking, falling, dropping, sliding, turning over while
sleeping, or other body movements) of the detection object U (e.g.,
elderly, patient, child or caregiver) in the field DR. However, the
specific frequency range does not include a frequency range
corresponding to a slow movement of the detection object U. The
slow movement may be, for example, the displacement change of the
detection object U on the surface of the chest cavity caused by
breathing or heartbeat. For ease of description, FIG. 2 only shows
one single detection object U, but the number of detection objects
in the field of the disclosure is not limited to FIG. 2.
[0019] In step S140, the processor 220 calculates a first
maintenance time length corresponding to an energy value of the raw
data signal SRD continuously being less than a first preset energy
value. In step S150, in response to the first maintenance time
length being greater than or equal to a first preset time length,
the processor 220 determines that a first event occurs in the
field. When a first maintenance time length TL1 (generated by the
energy value of the raw data signal SRD continuously being less
than the first preset energy value) is greater than or equal to the
first preset time length, the processor 220 determines that a first
event EV1 occurs in the field DR. The first event EV1 is an event
in which a fast motion does not occur in the field DR within the
first preset time length. In step S150, the processor 220 provides
the first alarm ALM1 corresponding to the first event EV1.
[0020] For example, the first preset energy value is 35 units, such
as 35 watts (but the disclosure is not limited thereto). The first
preset time length is, for example, 30 minutes. 30 minutes is
greater than or equal to the period of limb movement during sleep.
The first preset time length may be adjusted according to the
actual application situation, and the first preset time length in
the disclosure is not limited to the above. In step S130, the
processor 220 determines whether the energy value of the raw data
signal SRD is less than or equal to 35 watts. When the energy value
of the raw data signal SRD is less than or equal to 35 watts, the
processor 220 times a maintenance time during which the energy
value of the raw data signal SRD is less than or equal to 35 watts
to generate the first maintenance time length TL1. When the energy
value of the raw data signal SRD is greater than 35 watts, the
processor 220 resets the first maintenance time length TL1. On the
other hand, when the first maintenance time length TL1 is
determined to be greater than or equal to 30 minutes, the processor
220 determines that the first event EV1 occurs in the field DR. The
first event EV1 means that the fast motion does not occur in the
field DR within 30 minutes. In other words, the first event EV1
means that there is no motion such as walking, turning over while
sleeping, or other physical motions in the field DR for over 30
minutes. Accordingly, the processor 220 provides the first alarm
ALM1.
[0021] When the detection object is having a slow motion, such as
paralysis, slow falling or slow sliding, since it is difficult to
determine that the detection object is in danger based on the slow
motion, a corresponding alarm cannot be provided promptly.
Therefore, when the caregiver of the detection object is not
on-site, the detection object having paralysis, slow falling or
slow sliding is unable to receive any assistance in time. It should
be noted here that, the first event EV1 may be an event in which
the detection object U in the field DR who is unable to move due to
faint, slow falling or paralysis while there is no one in the field
DR. The first event EV1 may also be that there is no one in the
field DR. Accordingly, the motion detection device 10 can determine
that the first event EV1 occurs in the field DR and provide the
first alarm ALM1.
[0022] Returning to step S130, the processor 220 processes the raw
data signal SRD to obtain the energy distribution of the raw data
signal SRD. In step S160, the processor 220 performs an analysis
according to the energy distribution of the raw data signal SRD, so
as to determine whether a second event EV2 occurs in the field DR
according to the energy distribution of the raw data signal SRD.
The second event EV2 is, for example, a fall event. In step S170,
in case of determining that the second event EV2 occurs in the
field DR, the processor 220 calculates a second maintenance time
length corresponding to the energy value of the raw data signal SRD
continuously being less than a second preset energy value. In step
S180, in response to a second maintenance time length TL2 being
greater than or equal to a second preset time length, the processor
220 provides the second alarm ALM2 corresponding to the second
event EV2.
[0023] For example, the second preset energy value is 35 units,
such as 35 watts (but the disclosure is not limited thereto). The
second preset time length is, for example, 5 seconds (but the
disclosure is not limited thereto). In step S160, in case of
determining that the second event EV2 occurs, the processor 220
determines whether the energy value of the raw data signal SRD is
less than or equal to 35 watts. When the energy value of the raw
data signal SRD is less than or equal to 35 watts, the processor
220 times a maintenance time during which the energy value of the
raw data signal SRD is less than or equal to 35 watts to generate
the second maintenance time length TL2. When the energy value of
the raw data signal SRD is greater than 35 watts, the processor 220
stops timing or resets the second maintenance time length TL2. When
the second sustain time length TL2 is determined to be greater than
or equal to 5 seconds, it means that the fast motion does not occur
in the field DR for over 5 seconds after the second event EV2
occurs. The above situation may be the fast motion caused by the
detection subject U being unable to move (struggling) after
falling, and there is no one in the field DR to provide assistance.
In this case, the processor 220 correspondingly provides the second
alarm ALM2.
[0024] In this way, the motion detection device 10 applying the
non-contact motion detection method can determine whether the
second event EV2 occurs in the field DR, determine whether the
detection object U is unable to move after falling in the field DR,
and provide the second alarm ALM2 if so. In addition, the motion
detection device 10 can also determine that the first event EV1
occurs in the field DR and provide the first alarm ALM1.
Accordingly, the motion detection device 10 applying the
non-contact motion detection method can detect the energy of the
field DR and the energy of the motion of the detection object U, so
as to improve the accuracy of the overall situation in the field DR
when the detection object U is in danger, and provide the correct
alarm.
[0025] In this embodiment, the processor 220 of this embodiment is,
for example, a central processing unit (CPU) or other programmable
devices for general purpose or special purpose such as a
microprocessor and a digital signal processor (DSP), micro control
unit (MCU), a programmable controller, an application specific
integrated circuit (ASIC), a programmable logic device (PLD) or a
combination of other similar devices, which can load in computer
programs for execution.
[0026] In this embodiment, the motion detection device 20 further
includes a communication interface 230. The communication interface
230 is coupled to or electrically connected to the processor 220.
In this embodiment, the processor 220 transmits the first alarm
ALM1 and the second alarm ALM2 to other devices, such as a central
control terminal or a cloud server, through the communication
interface 230. However, the disclosure is not limited in this
regard. The communication interface 230 receives a setting signal
(not shown) from the outside of the motion detection device 20, and
the processor 220 adjusts an internal setting of the processor 220
according to the setting signal. In another embodiment, the
processor 220 transmits one or more of the raw data signal SRD, a
partial processing result of the raw data signal SRD, the first
alarm ALM1 and the second alarm ALM2 to the other devices through
the communication interface 230. The communication interface 230
may be a wired communication interface such as Universal
Asynchronous Receiver Transmitter (UART)/Inter-Integrated Circuit
Bus (I2C)/Serial Peripheral Interface (SPI)/Controller Area Network
(CAN)/Recommended Standard (RS) 232/Recommended Standard (RS) 422
interfaces, and may also be a wireless communication interface such
as a wireless sensor network (e.g., EnOcean/Bluetooth/ZigBee), a
cellular network (2G/3G/Long Term Evolution Technology (LTE)/5G), a
wireless local area network (e.g., Wireless Local Area Network
(WLAN)/Global Interoperability for Microwave Connectivity (WiMAX)),
a short-distance point-to-point communication (e.g., Radio
Frequency Identification (RFID)/EnOcean/Near Field Communication
(NFC)) interfaces (but not limited the above).
[0027] Referring to FIG. 3, FIG. 3 is a schematic diagram
illustrating a motion detection device according to a second
embodiment of the disclosure. A motion detection device 30 includes
a detection circuit 310, a processor 320 and a communication
interface 330. The implementation details of the processor 320 and
the communication interface 330 can be sufficiently taught in the
embodiment of FIG. 2 and will not be repeated here. In this
embodiment, the detection circuit 310 is a continuous wave radar.
In other embodiments, the detection circuit 310 may be a detection
circuit of other types of millimeter wave radars (e.g., a frequency
modulated continuous wave (FMCW) radar or an ultra-wideband (UWB)
radar). In this embodiment, the detection circuit 310 includes an
analog-to-digital converter (ADC) 311, a transmitting antenna 312,
a receiving antenna 313, a mixer 314, an amplifier 315, an
oscillator 316, a digital-to-analog converter (DAC) 317, an
intermediate frequency (IF) amplifier 318 and an amplifier 319. The
digital-to-analog converter 317 is coupled to or electrically
connected to the processor 320 to receive digital detection data DT
provided by the processor 320. The digital-to-analog converter 317
generates an analog control voltage according to the digital
detection data DT. The oscillator 316 is coupled to or electrically
connected to the digital-to-analog converter 317. The oscillator
316 receives the analog control voltage and generates a detection
signal based on a carrier frequency. The amplifier 315 is coupled
to or electrically connected between the transmitting antenna 312
and the oscillator 316. The amplifier 315 gains the detection
signal and provides the gained detection signal to the transmitting
antenna 312. The transmitting antenna 312 transmits the detection
signal to a field. In this embodiment, the detection signal is a
radio frequency signal.
[0028] The receiving antenna 313 receives a reflection signal from
the field. In this embodiment, the reflection signal is a
reflection radio frequency signal. The amplifier 319 is a low noise
amplifier (LNA). The mixer 314 is coupled or electrically connected
to the amplifier 319 and the oscillator 316. The mixer 314 receives
the reflection radio frequency signal and the radio frequency
signal output by the oscillator 316, and correspondingly obtains a
mixed signal MS containing the Doppler component of the reflection
radio frequency signal. The intermediate frequency amplifier 318 is
coupled or electrically connected to the mixer 314. The
intermediate frequency amplifier 318 performs a filtering operation
on the mixed signal MS according to a specific frequency band. The
intermediate frequency amplifier 318 also gains the mixed signal MS
so as to convert the mixed signal MS into the raw data signal SRD.
The analog-to-digital converter 311 is coupled or electrically
connected between the intermediate frequency amplifier 318 and the
processor 220.
[0029] Referring to FIG. 5A to FIG. 5C and FIG. 6, FIG. 5A to FIG.
5C are flowcharts illustrating a non-contact motion detection
method according to the second embodiment of the disclosure. FIG. 6
is a schematic diagram illustrating a processor according to an
embodiment of the disclosure. In this embodiment, the processor 220
includes timers TC1 to TC3. The timers TC1 to TC3 are respectively
coupled or electrically connected to the processor 220. The
processor 220 can be adapted to serve as the motion detection
device 10 of FIG. 1. The implementation details in steps S210 and
S220 can be sufficiently taught by steps S110 and S120 in the first
embodiment, which are not be repeated here. In step S230, the
processor 220 converts data of the raw data signal SRD to obtain an
energy distribution and an energy value of the raw data signal SRD.
In this embodiment, the processor 220 performs a short-time Fourier
transform (STFT) on the data of the raw data signal SRD to obtain a
frequency-energy distribution of the raw data signal SRD in each
time interval. In step S230, the processor 220 obtains the energy
distribution and the energy value of a frequency range.
[0030] After step S230, the non-contact motion detection method of
this embodiment will enter step S260 and enter step S240 via a step
node A. In this embodiment, step S260 includes steps S261 to S264.
In step S261, the processor 220 determines whether the energy
distribution obtained in step S230 is a suspicious sample
corresponding to the second event EV2. If the processor 220
determines that the sample is the suspicious sample corresponding
to the second event EV2, the processor 220 will enter step S262 to
analyze the suspicious sample. In this embodiment, the processor
220 analyzes the energy distribution of the raw data signal SRD
according to a motion detection algorithm. The motion detection
algorithm is trained by an artificial neural network. The processor
220 may, for example, analyze the suspicious samples by using a
machine learning model (but the disclosure is not limited thereto).
On the other hand, if determining that the sample is not the
suspicious sample corresponding to the second event EV2, the
processor 220 returns to step S210. The machine learning model may
be a Long Short-Term Memory (LSTM) model, a Recurrent Neural
Networks (RNN) model, a Convolutional Neural Networks (CNN) model,
and a Deep Neural Network (DNN) model, or a Region-based
Convolutional Neural Networks (R-CNN) model.
[0031] In step S263, the processor 220 determines whether an energy
distribution of the suspicious sample matches an energy
distribution of the second event EV2. If the energy distribution of
the suspicious sample matches the energy distribution of the second
event EV2, the processor 220 determines that the second event EV2
occurs in the field (the field DR shown in FIG. 1) in step S264.
Then, the non-contact motion detection method of this embodiment
enters step S270 via a step node B. On the other hand, if the
energy distribution of the suspicious sample does not match to the
energy distribution of the second event EV2, the process returns to
step S210.
[0032] Here, an example is provided below to illustrate the
implementation details of steps S230 and S260. Referring to FIG. 1,
FIG. 5A and FIG. 7, FIG. 7 is a schematic diagram illustrating a
conversion of data of a reflection signal into an energy window
slot according to an embodiment of the disclosure. The schematic
diagram of FIG. 7 may correspond to step S230. In step S230, the
processor 220 receives the data of the raw data signal SRD over
time and converts the data of the raw data signal SRD into one
single spectrum unit by a unit time length. In this embodiment, the
unit time length is, for example, 0.128 seconds. Accordingly, a
spectrum unit U01 shows an energy distribution of 0 to 250 Hz in 0
to 0.128 seconds. A spectrum unit U02 shows an energy distribution
of 0 to 250 Hz in 0.129 to 0.256 seconds, and the rest can be
deduced by analogy. In this embodiment, a segment of 25 consecutive
spectrum units U01 to U25 is used for description.
[0033] The processor 220 extracts a part of the energy distribution
in the spectrum units U01 to U25 based on a frequency range. The
frequency range is 100 to 250 Hz. In this embodiment, the processor
220 can filter out the energy distribution below 100 Hz and retain
the energy distribution between 100 and 250 Hz. In this way, the
energy distribution extracted by the processor 220 includes an
energy spectrum generated by the fast motion of the detection
object U and excludes an energy spectrum generated by the slow
motion of the detection object U. In this way, the processor 220
can reduce the amount of subsequent calculations to save
calculation resources. The processor 220 combines multiple adjacent
spectrum units into one single energy window slot. For example, the
processor 220 combines the spectrum units U01 to U05 into an energy
window slot WS1, combines the spectrum units U06 to U10 into an
energy window slot WS2, and so on and so forth. Based on this, the
processor 220 generates the energy window slots WS1 to WS5. The
energy window slots WS1 to WS5 respectively represent an energy
distribution change in the frequency range of 100 to 250 Hz in 0.64
seconds. The processor 220 then combines the energy window slots
WS1 to WS5 into a sample signal SP(t). A time length of the sample
signal SP(t) is 3.2 seconds. Since a time period of a complete fall
motion is approximately 2 to 3 seconds, the time length of the
sample signal SP(t) is sufficient to detect a complete motion of a
suspected falling.
[0034] Next, when a spectrum unit U26 is generated, an energy
window slot WS1' is generated according to the spectrum units U02
to U06. WS2' is generated based on spectrum units U07 to U11, and
the rest can be deduced by analogy. Accordingly, the processor 220
generates the energy window slots WS1' to WS5', and combines the
energy panes WS1' to WS5' into a sample signal SP(t+0.128).
[0035] In step S261, the processor 220 determines whether an energy
intensity of the middle energy window slot WS3 in the sample signal
SP(t) is greater than the energy intensities of the other energy
window slots WS1, WS2, WS4 and WS5. When the energy intensity of
the energy window slot WS3 is determined to be greater than the
energy intensities of the energy window slots WS1, WS2, WS4 and
WS5, the sample signal SP(t) is determined as the suspicious
sample. The non-contact motion detection method includes then
enters step S262. On the other hand, when the energy intensity of
the energy window slot WS3 is determined to be less than or equal
to the energy intensities of the energy window slots WS1, WS2, WS4
and WS5, the non-contact motion detection method returns to step
S210.
[0036] Referring back to the embodiment of FIGS. 5A to 5C and FIG.
6, the processor 220 calculates the first maintenance time length
TL1 in step S240 in this embodiment. In step S240, the processor
220 uses the processor 220 to determine whether the energy value of
the raw data signal SRD is less than or equal to the first preset
energy value. When the energy value of the raw data signal SRD is
determined to be less than or equal to the first preset energy
value, the processor 220 instructs the timer TC1 to accumulate a
maintenance time during which the energy value of the raw data
signal SRD is less than or equal to the first preset energy value,
so as to calculate the first maintenance time length TL1. Step S250
includes steps S251 to S255. In step S251, the processor 220
determines whether the energy value of the raw data signal SRD is
greater than the first preset energy value. When the energy value
of the raw data signal SRD is determined to be greater than the
first preset energy value, the processor 220 instructs the timer
TC1 to reset the first maintenance time length TL1 in step S252,
and executes step S240. That is, because the energy value of the
raw data signal SRD is greater than the first preset energy value,
the operation of the timer TC1 will be interrupted in step S252,
and the timing will be restarted in step S240. On the other hand,
when the energy value of the raw data signal SRD is determined to
be still less than or equal to the first preset energy value, the
processor 220 further determines whether the first maintenance time
length TL1 reaches the first preset time length in step S253.
[0037] In step S253, when the first maintenance time length TL1 is
determined to be less than the first preset time length, the
processor 220 performs a step loop of steps S240, S251 and S253.
When the first maintenance time length TL1 is determined to be
greater than or equal to the first preset time length, the
processor 220 determines that the first event EV1 occurs in the
field in step S254, and provides the first alarm ALM1 corresponding
to the first event EV1 in step S255.
[0038] In step S270, the processor 220 determines whether the
energy value of the raw data signal SRD is less than or equal to
the second preset energy value. When the energy value of the raw
data signal SRD is determined to be less than or equal to the
second preset energy value, the processor 220 instructs the timer
TC2 to accumulate a maintenance time during which the energy value
of the raw data signal SRD is less than or equal to the second
preset energy value, so as to calculate the second maintenance time
length TL2. Step S280 includes steps S281 to S287. In step S281,
the processor 220 determines whether the energy value of the raw
data signal SRD is greater than the second preset energy value.
When the energy value of the raw data signal SRD is determined to
be still less than or equal to the second preset energy value, the
processor 220 further determines whether the second maintenance
time length TL2 reaches the second preset time length in step
S282.
[0039] In step S282, the processor 220 determines whether the
second maintenance time length TL2 is less than the second preset
time length (e.g., 5 seconds, but the disclosure is not limited
thereto). When the second maintenance time length TL2 is determined
to be less than the second preset time length, the processor 220
performs a step loop of steps S270, S281 and S282. On the other
hand, when the second maintenance time length TL2 is determined to
be greater than or equal to the first preset time length, the
processor 220 determines that the fast motion does not occur over
the first preset time length after the second event EV2 occurs in
the field. The above situation may be caused by the detection
subject being unable to move (struggling) after falling, and there
is no one in the field to provide assistance. Accordingly, the
processor 220 provides the second alarm ALM2 in step S283.
[0040] Returning to step S281, when the energy value of the raw
data signal SRD is determined to be greater than the second preset
energy value, the processor 220 instructs the timer TC3 to
accumulate a maintenance time during which the energy value of the
raw data signal SRD is greater than the second preset energy value
in step S284, so as to generate a third maintenance time length
TL3. In step S284, the timer TC2 suspends timing. When the energy
value of the raw data signal SRD is determined to be greater than
the second preset energy value, it means that the fact motion
occurs within 5 seconds after the second event EV2 occurs (i.e.,
the second maintenance time length TL2 is less than the second
preset time length). The above situation may be caused by the
detection subject starting to move (struggling) after falling, and
there is someone in the field providing assistance. In step S284,
if the energy value of the raw data signal SRD is greater than the
second preset energy value, the timer TC3 suspends timing.
Accordingly, the third maintenance time length TL3 will not be
accumulated.
[0041] In step S285, the processor 220 determines whether the third
maintenance time length TL3 is less than a third preset time length
(e.g., 10 seconds, but the disclosure is not limited thereto). When
the third maintenance time length TL3 is determined to be less than
the third preset time length, the processor 220 instructs the timer
TC3 to reset the third maintenance time length TL3 in step S286,
and returns to step S270. On the other hand, when the third
maintenance time length TL3 is determined to be greater than or
equal to the third preset time length, it means that the fast
motion of up to 10 seconds occurs after the second event EV2
occurs. The second event EV2 is then determined as being properly
handled. Therefore, the processor 220 instructs the timers TC2 and
TC3 to end timing in step S287, and returns to step S210.
[0042] An embodiment of the disclosure further discloses an
emergency detection method. The emergency detection method can be
applied to the motion detection devices 10, 20 and 30 shown in FIG.
1 to FIG. 3. The emergency detection method includes: receiving a
reflection radio frequency signal (the reflection signal SR shown
in FIG. 1 to FIG. 3), and processing the reflection radio frequency
signal to obtain a raw data signal (the raw data signal SRD shown
in FIG. 1 to FIG. 3); processing the raw data signal to obtain an
energy response corresponding to a field, and determining a first
detection result according to the energy response; in response to
the energy response matching a first event (e.g., the first event
EV1 shown in FIG. 2), determining whether the first detection
result belongs to an emergency; processing the raw data signal
according to a motion detection algorithm to obtain a second
detection result; and in response to the second detection result
matching an abnormal motion (e.g., falling, sliding, collision or
the second event EV2 shown in FIG. 2), determining whether the
second detection result belongs to the emergency. For example, the
emergency is: (1) the fast motion does not occur for over the first
preset time period, or (2) after a first motion occurs, the fast
motion does not occur for over the second preset time period.
[0043] In summary, the disclosure receives the reflection signal
from the field, processes the reflection signal to obtain the
reflection signal, and calculates the first maintenance time length
corresponding to the energy of the reflection signal continuously
being less than the first preset energy value. When the first
maintenance time length is greater than or equal to the first
preset time length, the disclosure determines that the first event
occurs in the field, and provides the first alarm corresponding to
the first event. In case of determining that the second event
occurs in the field, the disclosure further calculates the second
maintenance time length corresponding to the energy value of the
raw data signal continuously being less than the second preset
energy value. When the second maintenance time length is greater
than or equal to the second preset time length, the disclosure
provides the second alarm corresponding to the second event. In
this way, the disclosure can detect the energy generated by the
motion of the detection object in the sensing field, so as to
improve the accuracy of the overall situation in the field when the
detection object is in danger, and provide the correct alarm.
* * * * *